Retirement strategies need new ways of judging the future
The bloody debate over Your Future Your Super (YFYS) shows the difficulty in assessing even relatively simple accumulation products. Assessing retirement strategies – absent the decades of data from the period in which they’re actually being put to work – will be even harder. To that end, ANU professor Geoff Warren (photo at top) and David Bell of the Conexus Institute have developed an assessment framework that could be put to work by super funds, asset consultants, research houses and regulators.
“The outcome you’re supposed to deliver is an income stream,” Warren says. “That could happen over twenty or thirty years and you can’t wait to see the outcomes to work out whether it’s working. And investment performance is only really a small part of the piece here. You have to look at if they’re delivering an income stream in a suitable way for the member.”
“The overriding message is that the only way you can do that is to answer the question of whether fund trustees are doing the best job they possibly can to deliver the right outcomes to the right members. And that’s an ex ante (forward looking) analysis.”
Looking back on a single income stream won’t tell you whether trustees did the right job of putting it together in the first place – there could be a slew of reasons for that outcome, including market movements, members living longer than expected, dying early, or doing something that the trustee didn’t recommend.
What Warren and Co. propose is a check list, supported by a stochastic quantitative model. The checklist would cover whether funds have the right governance structures in place, the right people to manage their retirement programs, whether they’re offering the right strategies and products to members through them, whether they’re cohorting those members in the right way, and whether they’re providing guidance through advice, calculators, and engagement.
The quant model then simulates “perpetual income outcomes”; a fund might then run its retirement strategy against a baseline 60/40 account-based pension with minimum drawdown rules. What Warren and Bell found while they were doing their research was that plenty of funds, asset consultants and regulators were already building them.
“You might find that they tweak the drawdown strategy to draw a bit more out than the minimum, and you find that income goes up and the balance at death goes down and there will be a better income outcome,” Warren says. “So tick, you’ve made an improvement.”
“Then they could ask ‘what if we do a smarter drawdown rule, one that responds to changes in market returns?. That might manage the risk of running out of money, and the quant model will tell you that’s a bit better still, so you’ve made another improvement.”
“The checklist covers everything they should be doing, including the guidance, as a way of analysing the outcomes you can expect from the strategy. The quant model would help and tell funds if when they made changes or tweaks they delivered better member outcomes. They would use both for improvement.”
What spurred Warren and Co. to put the research together were signs that regulators and policymakers were considering a YFYS-style measurement of retirement strategies – a fraught proposition, given they won’t have the luxury of looking back over the decades-long period where those retirement strategies are implemented. It’s unlikely that will now come to pass, but questions remain on how to assess the most important part of superannuation.
“I’m not sure there’s an easy way out with retirement,” Warren says. “It’s too complex. It’s complex because the problem is complex and the problem is complex because members are so different. This is one of the things the industry is grappling with now. Accumulation is about making the balance as big as possible before they get to retirement and it’s largely about returns. In retirement it’s about a whole bunch of other, different things… That’s why the YFYS model won’t apply.”